Geneva, Switzerland — A coalition of independent fraud practitioners, operating under the provisional name the International Brotherhood of Authentic Deception (IBAD), has issued a joint statement condemning the accelerating deployment of artificial intelligence in what the coalition characterizes as their occupational domain, warning that unchecked automation threatens to eradicate what one signatory called “a practice requiring genuine human presence, earned instinct, and the irreplaceable element of emotional risk.” The statement, distributed simultaneously across seventeen encrypted channels, requests formal recognition of the industry's labor interests and calls on regulators to distinguish, legally and commercially, between artisanal and algorithmic fraud.
The document, which runs to forty-three pages including seven appendices and a glossary of contested terminology, frames the current moment as a crisis of craft displacement no different in moral character from the automation anxiety that has periodically restructured every other skilled trade since the industrial revolution. That the trade in question is confidence fraud, elder scamming, romance manipulation, and investment deception does not appear to have introduced hesitation into the coalition's advocacy posture.
"Scamming is and should remain a human endeavor," the statement reads in its opening declaration. "We will not allow machines to replace the sanctity of scamming."
The phrase "sanctity of scamming" appears eleven times in the document and is treated throughout as self-evidently meaningful.
The Formation of the Coalition
IBAD traces its founding to a virtual summit held earlier this year, convened after a series of internal discussions among senior practitioners who noticed that their close rates had begun declining in ways that did not correlate with victim sophistication, law enforcement activity, or economic conditions. The decline, several members reported independently, appeared to correlate specifically with periods of expanded AI chatbot deployment.
One founding member, identified in coalition materials only as "V.K., Lagos Chapter," described the moment of recognition with evident frustration: the phishing campaigns he had refined over eleven years — calibrated by region, adjusted by season, personalized by research — were being undercut in response rate by systems that had been operating for three months and had never spoken to a human being. Potential victims were not becoming more resistant to fraud. They were simply being reached first, and more efficiently, by automated systems that could operate continuously across time zones without fatigue, negotiation costs, or moral qualms.
"A machine does not care whether the person on the other end is frightened," V.K. reportedly told the summit. "I have always cared. Not in a way that stopped me, but in a way that made me better at it."
Whether this distinction constitutes an ethical argument or merely a professional one was not resolved at the summit and is not resolved in the statement.
The Argument from Craft
The central thesis of IBAD's statement is what coalition drafters call the Craft Preservation Argument, which holds that traditional fraud is a skilled practice with genuine technical requirements, a meaningful learning curve, and qualitative properties that automation cannot reproduce. The argument is presented not as a defense of the practice's ethics but as a description of its nature — the coalition is explicit that it is not asking anyone to approve of what its members do, only to recognize that they do it in ways that differ materially from what machines do.
"Where is the artistry?" asks the statement's second section, in language that reads less like advocacy and more like genuine grievance. "The improvisation? The hustle?"
These are, the coalition argues, not rhetorical questions. The practice of confidence fraud — the word "fraud" appears nowhere in the document, which prefers "persuasion-adjacent engagement" — requires its practitioners to model human psychology in real time, adjust to resistance without triggering flight responses, construct and maintain internally consistent false identities across extended interactions, recognize and exploit emotional windows that close in seconds, and accept personal risk of exposure, arrest, and consequence. These requirements, IBAD contends, constitute the conditions of skilled practice. Removing them is not an improvement to the process. It is the elimination of the process and its replacement with something categorically different.
The statement draws an analogy to financial markets, where algorithmic trading displaced floor traders not because it was better at finance but because it was faster at execution, eliminating in the process the human judgment that floor traders believed, perhaps self-servingly but not entirely incorrectly, constituted the value of their role. IBAD suggests the same substitution is occurring in fraud, and that the victims of this substitution are, paradoxically, both the practitioners being displaced and the potential fraud targets, who now face a vastly larger volume of attempts even as each individual attempt becomes technically less sophisticated.
"Scale was never the virtue," the statement notes. "The virtue, if you could call it that, was contact."
Proposed Regulatory Framework
The statement's third section, titled "Toward a Certification Economy for Authentic Persuasion-Adjacent Engagement," outlines a proposed regulatory structure that the coalition acknowledges will require "creative jurisdictional interpretation" to implement. The proposals are presented with the seriousness of a professional licensing board submission.
The centerpiece is a certification system for what the document calls "authentic practitioners" — individuals who can demonstrate, through examination and peer review, that their fraud operations are conducted entirely by humans without AI-assisted content generation, automated targeting, or algorithmic personalization. Certification would be renewed annually and would entitle holders to use a proposed mark: a small stamp reading "Handcrafted Deception," rendered, in the appendix mockup, in a typeface chosen to evoke artisan food packaging.
The certification body, tentatively named the Institute for Deception Craft Standards (IDCS), would be governed by a board of senior practitioners and would establish minimum competency requirements including demonstrated proficiency in emotional manipulation techniques, a minimum of three years of active practice, and the successful completion of what the document describes as "a live assessment conducted against a consenting adult volunteer under controlled conditions." The document does not specify how volunteers would be recruited for this assessment, noting only that the logistics "remain under development."
Alongside certification, IBAD proposes "human verification requirements" for fraud operations above a certain scale — specifically, that any operation targeting more than 847 individuals simultaneously would be required to demonstrate that each contact was initiated by a human being. The figure of 847 is not explained in the text but appears with enough regularity across the document's appendices to suggest it carries significance within the coalition's internal deliberations.
The final proposal is a labeling standard for fraudulent communications: a requirement that AI-generated scam content be marked as such, in the same way that some jurisdictions require disclosure of AI-generated advertising. The coalition presents this proposal with apparent sincerity, arguing that victims deserve to know whether they are being deceived by a human or a machine, and that this distinction has "informational value independent of the outcome." The argument is not that labeled AI fraud is acceptable. The argument is that unlabeled AI fraud is worse.
Legal analysts who reviewed the proposals for this publication declined, uniformly, to comment on their enforceability. Several laughed audibly before declining.
The Academic Reception
The statement did not pass without academic notice. Dr. Henry Gutenberg of the Port-au-Prince Institute for Market Dysfunction, whose work on what he terms "the rationalization architectures of extractive industries" has been consistently cited and consistently ignored in equal measure, issued a response that circulated widely in labor economics departments before being quietly removed from two institutional repositories for reasons that were not explained.
Gutenberg's response, titled "Craft Discourse and the Displaced Predator: Labor Identity in the Automation of Harm," argues that IBAD's statement is analytically interesting precisely because it is structurally identical to every previous craft-displacement grievance in industrial history — the same language, the same appeals to embodied knowledge, the same distinction between artisanal quality and industrial volume — applied to a practice whose social harm is universally acknowledged. What makes the document revealing, Gutenberg writes, is not that it is dishonest but that it is completely honest: it makes the craft argument sincerely, without apparent awareness that the craft argument has always been made sincerely, and has always been worth examining on its own terms, regardless of what the craft produces.
"The weavers displaced by the Jacquard loom were not wrong that something was lost," Gutenberg wrote. "They were wrong only in their assessment of what that something was worth preserving. IBAD presents us with the same claim in a context that makes the underlying logic visible. The argument for craft is always an argument for a particular kind of human labor. It does not, and cannot, answer the question of whether that labor should exist."
Gutenberg noted that IBAD's statement had nonetheless identified a genuine phenomenon: AI-driven fraud does scale deception beyond what human practitioners can achieve, does reduce the per-unit cost of an attempt to near zero, and does remove from the process whatever feedback mechanisms — however perverse — human practitioners develop through sustained contact with their targets. Whether this makes AI fraud worse than human fraud in aggregate impact, he concluded, is an empirical question that existing research does not yet fully answer. The honest answer, Gutenberg suggested, is that it probably makes it worse in scale while potentially making it worse in different ways in terms of precision. Both conclusions are bad.
His institute issued a press release the following week announcing the launch of a three-year research program on the topic. Funding has not yet been secured. Several funding bodies contacted for this article declined to confirm whether they had received an application.
The Cybersecurity Industry Responds
Cybersecurity researchers and consumer protection advocates responded to the statement with a combination of amusement and genuine concern — not at IBAD's proposals, which they uniformly dismissed as unenforceable theater, but at the data underlying the coalition's complaints, which they found, on examination, to be substantially accurate.
Reports from multiple fraud monitoring organizations confirm that AI-assisted scam campaigns have expanded in volume by orders of magnitude over the preceding two years, while the per-attempt cost has declined to the point where campaigns targeting millions of individuals simultaneously have become financially viable for actors who would previously have lacked the resources to attempt them. The sophistication of individual attempts varies widely — some AI-generated fraud content remains easily detectable — but the detection problem has changed in character: the challenge is no longer identifying bad actors from a manageable pool of attempts but processing a volume of attempts that exceeds the capacity of any human-review system.
"The issue is not who does it," said one analyst at a major cybersecurity firm, who asked not to be identified because her employer had not authorized comment. "It's that it's happening."
The analyst noted, with some apparent frustration, that IBAD's statement had correctly identified several dynamics that her industry had been attempting to communicate to policymakers for years without generating comparable attention. The volume problem, the personalization problem, the time-zone-agnostic operation problem — these had all been raised in congressional testimony, in regulatory comment periods, in published research. The IBAD statement, by raising the same issues from the perspective of people being harmed by the automation rather than people being victimized by the practice, had in a single afternoon generated more public discussion than several years of conventional advocacy.
She declined to say whether she found this ironic. "Irony isn't in my job description," she said, and terminated the call.
The Internal Schism
Not everyone within IBAD endorsed the statement's position. A dissenting faction, characterized in coalition communications as "the modernization wing" and characterized in dissenting communications as "people who understand arithmetic," argued that the coalition's anti-automation stance was strategically incoherent and practically self-defeating.
The modernization wing is understood to consist primarily of younger practitioners and those operating in markets where AI-tool adoption has been fastest. Their argument, circulated in a counter-document titled "A Realist's Position on Operational Evolution," runs as follows: the automation of fraud is happening regardless of whether IBAD endorses it; the practitioners who survive will be those who integrate AI tools most effectively; and the pretense that craft resistance constitutes a viable competitive strategy reflects not principle but a failure to accept the conditions of the current operating environment.
"We are not ceramicists," reads the counter-document's opening line, in a register notably less elevated than IBAD's primary statement. "No one is going to pay a premium for hand-thrown scam emails."
The counter-document argues that the coalition's energy would be better spent developing internal standards for AI-assisted operations — ethical AI use guidelines, in the language the document borrows without apparent self-awareness from the technology industry — rather than attempting to preserve a craft model whose economic rationale is evaporating. It identifies three specific areas where AI tools provide irreplaceable efficiency gains: initial target identification, first-contact message generation, and response processing at scale. Human judgment, the document concedes, remains valuable in "high-value relationship cultivation" — extended con operations requiring sustained persona maintenance — but even here, AI tools can handle what the document calls "the administrative burden of early relationship stages."
The primary statement's authors responded to this document in an appendix added shortly before final distribution. The response is three sentences long. The third sentence reads: "This position represents exactly what we are fighting."
The internal vote on the statement passed, coalition sources report, with twenty-three in favor, seven opposed, and four abstentions from members who stated they needed more time to "assess the long-term strategic implications." The four abstaining members have not yet reported back. Coalition leadership has not followed up.
International Dimensions
The statement's reception varied significantly by geography, in ways that reveal the uneven distribution of both fraud operations and AI-tool access across global markets.
In West Africa, where organized fraud operations have historically employed thousands of people in what some economists have documented as a significant informal employment sector, the statement generated substantial discussion. Several regional practitioners interviewed for this article expressed sympathy with IBAD's position while doubting its effectiveness. The concern in these markets is not primarily ideological but economic: AI tools reduce the labor intensity of fraud operations, which reduces employment in sectors that, in some communities, represent the primary available income source for educated young adults facing formal labor market exclusion. The labor displacement argument, in this context, is not ironic. It is serious, and several practitioners made it without evident embarrassment about the nature of the work being displaced.
In Eastern Europe, where technically sophisticated fraud operations have increasingly integrated AI tools as they became available, the statement was received largely as nostalgic posturing. One practitioner, reached through intermediaries and speaking on condition of complete anonymity, described IBAD's position as "romanticism from people who haven't updated their methods since 2018." He reported that his operation had integrated AI-assisted content generation eighteen months earlier, had seen a 340% increase in throughput, and had experienced no discernible decline in per-unit success rate. He wished IBAD well and ended the conversation.
In Southeast Asia, practitioners familiar with the large-scale fraud compounds that have generated international law enforcement concern in recent years expressed the most ambivalent response. Operations at that scale had never been craft enterprises — they were, by design, industrial. The IBAD statement's framing, several respondents suggested, reflected the assumptions of small-scale independent operators and had limited relevance to the organizational models that actually dominated the industry's volume output. The coalition was, in this reading, not representing the industry but representing a particular stratum of the industry that was being rendered economically marginal by forces larger than AI adoption alone.
IBAD did not respond to requests for comment on its membership composition.
The Labeling Proposal and Its Peculiar Logic
Of all IBAD's proposals, the AI-labeling requirement has attracted the most sustained outside attention — not because analysts consider it viable but because its internal logic is genuinely difficult to dismiss.
The argument runs as follows: informed consent is a value that applies even in adversarial contexts; potential victims of fraud retain an interest in understanding the nature of the communication they are receiving; knowing that a message was generated by an AI rather than written by a person who had researched their specific circumstances and invested real time in the attempt represents information that could affect their response; therefore, labeling AI fraud content as AI-generated would serve a legitimate disclosure function, even if it cannot serve a protective one.
Consumer protection scholars who reviewed the argument for this publication were divided on whether to engage with it. Those who did engage noted that the logic was not entirely wrong — disclosure requirements do, in theory, serve informational functions independent of behavioral outcomes — but that the proposal suffered from an obvious implementation problem. Labeling requirements only function when the labeled party has an interest in compliance or faces credible enforcement consequences. Neither condition applies to fraud operators, who are, by definition, not engaged in compliance-oriented behavior.
"They are asking the people who make a living lying to disclose honestly that they are lying," said Professor Miriam Osei-Bonsu of the Consumer Finance Protection Research Collaborative, which is affiliated with an institution she declined to name. "I appreciate the recursive quality of this proposal. I do not think it will work."
IBAD's drafters appear to have anticipated this objection. A footnote to the labeling proposal notes that enforcement would be "challenging in the traditional sense" but argues that market pressures might achieve what regulation cannot: if human-authored fraud develops a reputation for higher quality — more personalized, more convincing, more contextually appropriate — a premium could emerge, creating voluntary incentives for authentic practitioners to differentiate their product. The footnote uses the phrase "product" without visible discomfort.
The proposal, taken seriously for a moment, describes a future in which fraud victims, given the option, might prefer to be deceived by a human. Whether this preference would be rational depends on assumptions about what "better fraud" means for the person being defrauded. IBAD does not address this question. Possibly they considered it impolite.
What the Coalition Is Actually Describing
Read charitably — and IBAD's document rewards charitable reading, in the way that documents do when their authors are describing something real about an experience they are unable to describe honestly — the statement is a record of an industry confronting the same automation dynamics that have restructured manufacturing, logistics, customer service, legal research, financial analysis, and creative production over the preceding decade. The experience of watching a machine replicate a skilled practice at scale and at speed, with no apparent understanding of what the practice means to the person who does it, is not unique to fraud. It is the characteristic experience of contemporary automation. IBAD's practitioners are having it in a context that forecloses sympathy, but they are having it.
What they are describing, stripped of its occupational context, is the loss of what labor economists call "tacit knowledge" — the embodied, context-dependent, relationship-mediated understanding that distinguishes expert practice from competent execution. In fraud, this knowledge is mostly harmful. But its replacement by systems that can execute without knowledge while scaling without limit is not obviously a net improvement for anyone outside the small group of people who profit from the automation.
"We built this," the statement concludes. "We will protect it."
The claim to having built "this" — meaning, apparently, the contemporary landscape of mass fraud — is historically complicated and analytically muddled. But the desire to protect it is, in its way, a recognizable human response to a machine doing your job in front of you.
At press time, both human and AI-driven scams were ongoing. Competing. For attention.
The Bottom Line
IBAD's statement is, in the most literal sense, a labor organization using the language of craft preservation to protect its members from technological unemployment. The fact that the labor in question is fraud does not make the underlying dynamics less real. Automation is eliminating the economic rationale for human-operated scam enterprises in the same way it has eliminated the economic rationale for dozens of other skilled practices, and the people being eliminated are experiencing this in the ordinary ways: with grievance, with nostalgia, with appeals to quality, and with proposals for regulatory protection that will not be implemented.
The statement is worth taking seriously precisely because it cannot be taken seriously. Its proposals are unenforceable, its regulatory aspirations are absurd, and its certification scheme would require defrauding a regulator in order to receive regulatory recognition. But the phenomenon it describes — mass automation of deception, operating at a scale that overwhelms any individual or institutional capacity to detect and resist — is not absurd. It is the actual situation. IBAD has described the actual situation from an angle that makes visible what years of cybersecurity advocacy could not: that the automation of harm is a labor story as much as a technology story, and that the labor story is being lived, in recognizable human terms, by people whose interests we are not obligated to protect.
The question the statement cannot ask — because asking it would require acknowledging that the answer is bad — is what it means that the most visible critics of automated fraud are the people being replaced by it. Every other stakeholder — victims, regulators, the cybersecurity industry, the public — has been making this argument for years without traction. IBAD made it once, from the wrong side of the table, and it generated a news cycle. The system found the story legible when its subjects were losers rather than victims. That is not a fact about IBAD. That is a fact about the system.
Editor's note: Following distribution of IBAD's statement, at least three consumer protection organizations issued responses noting that several of IBAD's empirical claims about AI fraud volume were accurate and citing them in ongoing regulatory proceedings. They did not cite the source. IBAD did not issue a follow-up. The AI scams continued. The human scams continued. No certification body was established. The "Handcrafted Deception" mark remains pending, in the sense that everything that has not yet happened remains pending.
¹ IBAD and the Artisanal Deception Protection Act are fictional constructs. The dynamics they describe — rapid AI-driven expansion of fraud volume, labor displacement in fraud-dependent informal economies, and the inadequacy of existing regulatory frameworks — are documented in published research from consumer protection agencies, academic institutions, and law enforcement bodies in multiple jurisdictions.
² Dr. Henry Gutenberg and the Port-au-Prince Institute for Market Dysfunction are recurring fictional constructs in this publication. His analysis is offered sincerely. It is not heeded. This is consistent with prior coverage.
³ The figure 847 appears in this article as a coalition-internal threshold. Its significance to the coalition was not explained. This publication has not investigated further. Some numbers are like that.
⁴ The "Handcrafted Deception" certification mark mockup described in this article was not reproduced here. Readers are invited to imagine it. It looks like what you think it looks like.